Table of Contents
Google has launched Gemini 3.7 Flash, its latest Flash-series AI model, with a focus on software engineering, web development, knowledge work and AI-agent workflows.
The release comes only three weeks after Gemini 3.6 Flash, showing how quickly Google is iterating on its workhorse models. Google describes Gemini 3.7 Flash as its most capable Flash model yet for coding and agents, with improvements driven by developer feedback and algorithmic changes.
The launch also comes with a major pricing incentive. Google is offering introductory API pricing of $0.75 per million input tokens and $3.75 per million output tokens through December 31, 2026. From January 1, 2027, the standard pricing is scheduled to rise to $1.50 per million input tokens and $7.50 per million output tokens.
Built for Coding and AI Agents
Gemini 3.7 Flash is designed for applications that require more than simple question answering.
Google says the model is better at debugging, resolving software issues and generating production-ready code. On DeepSWE v1.1, it scored 65.3%, compared with 49% for Gemini 3.6 Flash. On FrontierCode 1.1 Main, its score increased to 43.6% from 34.4%.
The model is also aimed at agentic workflows, where AI systems can plan tasks, use software tools and complete multiple steps with less human intervention.
Google says Gemini 3.7 Flash can better adapt when an agent encounters obstacles, clarify user intent when necessary and follow instructions more accurately.
Improvements Beyond Coding
The new model is not limited to programming.
Google says Gemini 3.7 Flash improves performance in knowledge-intensive areas such as finance, law and biosciences. Its GDP.pdf score increased to 34% from 22% on Gemini 3.6 Flash.
On AutomationBench, which evaluates real-world business workflow automation, the model scored 30.4%, compared with 17% for its predecessor.
Google also reports improvements in web development, including generating more functional layouts and feature-complete applications with fewer prompts.
Gemini Spark Gets the New Model
Google is also integrating Gemini 3.7 Flash into Gemini Spark, its subscription-based personal AI agent.
Spark is available to Google AI Pro and Ultra subscribers in more than 160 countries. With the new model, Google says Spark can perform knowledge-work tasks more efficiently, including working with Google Workspace applications.
For developers, the model is available through the Gemini API, Google AI Studio and Android Studio. Enterprise customers can access it through Google’s enterprise AI platforms.
Gemini 3.5 Pro Remains Delayed
Despite releasing another Flash model, Google has still not provided a release date for Gemini 3.5 Pro, its anticipated premium model.
Google previously said in July that Gemini 3.5 Pro was being tested with partners and would arrive “soon.” Thursday’s announcement did not provide a new timeline. Reuters reported that investors are closely watching the model’s progress as a measure of Google’s ability to compete with companies such as Anthropic and OpenAI.
The continued delay has led to speculation about Google’s longer-term model roadmap, including whether the company could eventually change its plans around Gemini 3.5 Pro. However, Google has not confirmed that the model has been cancelled or will be skipped.
Leadership Changes Add to the Pressure
The Gemini 3.7 Flash launch comes during significant changes at Google DeepMind.
Reuters reported that Google co-founder Sergey Brin has urged key AI employees to focus heavily on Gemini as Alphabet works to strengthen its position against rivals.
Meanwhile, Google DeepMind chief Demis Hassabis recently stepped aside in favor of deputy Koray Kavukcuoglu, while the two original technical co-leads of Gemini left the company to start a new venture.
These developments add another layer of uncertainty around Google’s AI strategy.
A Competitive Price Point
Pricing is one of the strongest features of Gemini 3.7 Flash.
The introductory $0.75 input and $3.75 output rates make the model substantially cheaper than many premium AI models. Google is using the lower price to encourage developers and businesses to build and operate production AI agents at scale.
However, the promotional pricing is temporary. Developers planning long-term deployments will need to account for the higher rates scheduled from January 2027.
FAQs
What is Gemini 3.7 Flash?
Gemini 3.7 Flash is Google’s latest Flash model, designed for coding, web development, knowledge work and agentic workflows.
How much does Gemini 3.7 Flash cost?
Introductory pricing is $0.75 per million input tokens and $3.75 per million output tokens through December 31, 2026.
Is Gemini 3.5 Pro available?
Google has not announced a confirmed public release date for Gemini 3.5 Pro.
Where can Gemini 3.7 Flash be used?
It is available through the Gemini API, Google AI Studio, Android Studio, enterprise platforms and Gemini Spark for eligible subscribers.
Conclusion
The launch of Gemini 3.7 Flash strengthens Google’s offering for developers and businesses building AI-powered software and autonomous agents. The model brings reported gains in coding, web development, document analysis and business automation while launching with significantly reduced introductory API pricing.
At the same time, the absence of a release date for Gemini 3.5 Pro keeps questions around Google’s flagship AI roadmap alive. For now, Google is relying on the rapid development of its Flash series while continuing work on its broader Gemini strategy.